ARTICLE IN PRESS Neurobiology of Learning and Memory xxx 2006 xxx xxx www elsevier com locate ynlme A daytime nap containing solely non REM sleep enhances declarative but not procedural memory Matthew A Tucker Yasutaka Hirota Erin J Wamsley Hiuyan Lau Annie Chaklader William Fishbein Laboratory of Cognitive Neuroscience and Sleep The City College of the City University of New York 138th Street and Convent Ave New York NY 10031 USA Received 25 January 2006 revised 4 March 2006 accepted 10 March 2006 Abstract The specialized role that sleep speci c brain physiology plays in memory processing is being rapidly clari ed with a greater understanding of the dynamic complex and exquisitely orchestrated brain state that emerges during sleep Behaviorally the facilitative role of non REM NREM sleep primarily slow wave sleep for declarative but not procedural memory performance in humans has been demonstrated in a number of nocturnal sleep studies However subjects in these studies were tested after periods of sleep that contained REM sleep in addition to NREM sleep and comparison wake groups were subjected to mild sleep deprivation To add some clarity to the ndings of these nocturnal studies we assessed performance on declarative and procedural memory tasks following a daytime training retest interval containing either a short nap that included NREM without REM sleep or wakefulness Consistent with previous ndings we show that after a comparatively brief sleep episode subjects that take a nap improve more on a declarative memory task than subjects that stay awake but that improvement on a procedural memory task is the same regardless of whether subjects take a nap or remain awake Slow wave sleep was the only sleep parameter to correlate positively with declarative memory improvement These ndings are discussed with reference to the general bene ts of napping and within the broader context of a growing literature suggesting a role for NREM speci c physiology for the processing of declarative memory 2006 Elsevier Inc All rights reserved Keywords Sleep Declarative memory Procedural memory Consolidation Daytime napping Slow wave sleep Hippocampus 1 Introduction After more than 40 years of research it has become well known that performance on procedural memory tasks e g visual discrimination mirror tracing etc is enhanced following periods of sleep containing rapid eye movement REM sleep and that performance on these tasks is impaired following REM sleep deprivation Fishbein Gutwein 1977 Karni Tanne Rubenstein Askenasy Sagi 1994 Mednick Nakayama Stickgold 2003 Smith 2001 Stickgold Whidbee Schirmer Patel Hobson 2000 In humans the evidence is much weaker that REM Corresponding author E mail address w shbein ccny cuny edu W Fishbein 1074 7427 see front matter 2006 Elsevier Inc All rights reserved doi 10 1016 j nlm 2006 03 005 sleep promotes processing of declarative hippocampus dependent memories i e memories for fact based information and personal events which has led some to abandon the idea that sleep speci c physiological processes play any role at all in the processing of declarative memories Siegel 2001 Vertes 2004 However a strong case has recently been made for the involvement of non REM NREM sleep Stages 2 4 and in particular slow wave sleep SWS Stages 3 and 4 in the processing of declarative memories A small number of behavioral studies have shown that periods of sleep dominated by SWS enhance declarative memory recall but do not a ect performance on procedural memory tasks Plihal Born 1997 1999a These studies are complemented by compelling neurophysiological hypotheses outlining the potential ARTICLE IN PRESS 2 M A Tucker et al Neurobiology of Learning and Memory xxx 2006 xxx xxx mnemonic mechanisms of SWS speci c declarative hippocampus related memory processing Buzsaki 1989 1998 Hasselmo 1999 Sejnowski Destexhe 2000 However despite these promising ndings there are some outstanding issues to be addressed to more rmly establish this relationship including the use of research designs that isolate the e ects of NREM sleep independent of REM sleep and the development of research designs that avoid sleep deprivation Of the handful of behavioral research designs examining the bene ts of sleep for memory processing in humans the early late sleep paradigm has consistently revealed a dissociation between memory system tested procedural versus declarative and stage of sleep SWS versus REM Barrett Ekstrand 1972 Plihal Born 1997 1999a Yaroush Sullivan Ekstrand 1971 By comparing performance after the rst 3 4 h of nocturnal sleep which is dominated by SWS with performance after the last 3 4 h of sleep during which REM sleep predominates it is possible to evaluate the relative contribution of these two stages of sleep to processing of di erent types of memory It should be noted that the amount of Stage 2 sleep is nearly equal in the early and late halves of the night Using this design it has been shown that early sleep facilitates recall of declarative information paired associates learning and spatial memory and that late sleep bene ts performance on procedural memory tasks mirror tracing Plihal Born 1997 These studies also demonstrate that subjects that sleep either early or late in the sleep period perform better than their counterparts that stay awake during these intervals This early late sleep design has been used to assess the relationship between sleep stage and emotional memory Wagner Gais Born 2001 sensory memory Gais Plihal Wagner Born 2000 and recognition memory Drosopoulos Wagner Born 2005 While the ndings of the above mentioned studies provide encouraging evidence that SWS facilitates declarative memory processing several unresolved issues remain Though it is true that the rst half of the sleep period contains increased amounts of SWS it still contains a substantial amount of REM sleep It is common for a normal sleeper to cycle through two NREM REM periods during this 3 4 h sleep interval which leaves open the question whether it is solely NREM sleep or the combination of NREM and REM sleep that produce the memory bene ts following sleep Based on a number of behavioral ndings it has been suggested that SWS and REM sleep play complementary roles in memory processing Ambrosini Giuditta 2001 Ficca Lombardo Rossi Salzarulo 2000 Gais et al 2000 Stickgold et al 2000 Also it is important to consider the striking di erences between REM sleep and SWS REM sleep is characterized by a desynchronized